On Tue, Mar 10, 2009 at 6:37 AM, Daniel Jacobowitz <drow@false.org> wrote:

> Disabling a shlib_event breakpoint is pretty weird - there won't be an
> event to restore it, so unless we're restarting the program it won't
> be enabled again ever.  If we're disabling a step-return or finish
> breakpoint, we'll lose control of the inferior.  And so forth...

Thanks, I think I understood how this works now.

Attached patch is simply a cleanup (i.e. I dont't really care if
it is applied or not, though I think it is more efficient and
easier to understand the source). It introduces no regressions
of Linux/x86_64.

OK to commit?
-- 
Paul Pluzhnikov


2009-03-11  Paul Pluzhnikov  <ppluzhnikov@google.com>

	    * breakpoint.c (disable_breakpoints_in_shlibs): Use
	    solib_contains_address_p instead of searching.